Integrated prediction of the effect of mutations on multiple protein characteristics

Site-directed mutagenesis is routinely used in modern biology to elucidate the functional or biophysical roles of protein residues, and plays an important role in the field of rational protein design. Over the past decade, a number of computational tools have been developed that can predict the effect of point mutations on a protein's biophysical characteristics. However, these programs usually provide predictions for only a single characteristic. Furthermore, online versions of these tools are often impractical to use for examination of large and diverse sets of mutants. We have created a new web application, (http://enzyme.ucd.ie/PEAT_SA), that can simultaneously predict the effect of mutations on stability, ligand affinity and pK(a) values. PEAT-SA also provides an expanded feature-set with respect to other online tools which includes the ability to obtain predictions for multiple mutants in one submission. As a result, researchers who use site-directed mutagenesis can access state-of-the-art protein design methods with a fraction of the effort previously required. The results of benchmarking PEAT-SA on standard test-sets demonstrate that its accuracy for all three prediction types compares well to currently available tools. We illustrate PEAT-SA's potential by using it to investigate the influence of mutations on the activity of Subtilisin BPN'. This example demonstrates how the ability to obtain a wide range of information from one source, that can be combined to obtain deeper insight into the influence of mutations, makes PEAT-SA a valuable service to both experimental and computational biologists.     

Conservation Weighting Functions Enable Covariance Analyses to Detect Functionally Important Amino Acids

The explosive growth in the number of protein sequences gives rise to the possibility of using the natural variation in sequences of homologous proteins to find residues that control different protein phenotypes. Because in many cases different phenotypes are each controlled by a group of residues, the mutations that separate one version of a phenotype from another will be correlated. Here we incorporate biological knowledge about protein phenotypes and their variability in the sequence alignment of interest into algorithms that detect correlated mutations, improving their ability to detect the residues that control those phenotypes. We demonstrate the power of this approach using simulations and recent experimental data. Applying these principles to the protein families encoded by Dscam and Protocadherin allows us to make testable predictions about the residues that dictate the specificity of molecular interactions.     

Comparing phylogeny and the predicted pathogenicity of protein variations reveals equal purifying selection across the global human mtDNA diversity

We used detailed phylogenetic trees for human mtDNA, combined with pathogenicity predictions for each amino acid change, to evaluate selection on mtDNA-encoded protein variants. Protein variants with high pathogenicity scores were significantly rarer in the older branches of the tree. Variants that have formed and survived multiple times in the human phylogenetics tree had significantly lower pathogenicity scores than those that only appear once in the tree. We compared the distribution of pathogenicity scores observed on the human phylogenetic tree to the distribution of all possible protein variations to define a measure of the effect of selection on these protein variations. The measured effect of selection increased exponentially with increasing pathogenicity score. We found no measurable difference in this measure of purifying selection in mtDNA across the global population, represented by the macrohaplogroups L, M, and N. We provide a list of all possible single amino acid variations for the human mtDNA-encoded proteins with their predicted pathogenicity scores and our measured selection effect as a tool for assessing novel protein variations that are often reported in patients with mitochondrial disease of unknown origin or for assessing somatic mutations acquired through aging or detected in tumors.     

Robust classification of protein variation using structural modelling and large-scale data integration

Existing methods for interpreting protein variation focus on annotating mutation pathogenicity rather than detailed interpretation of variant deleteriousness and frequently use only sequence-based or structure-based information. We present VIPUR, a computational framework that seamlessly integrates sequence analysis and structural modelling (using the Rosetta protein modelling suite) to identify and interpret deleterious protein variants. To train VIPUR, we collected 9477 protein variants with known effects on protein function from multiple organisms and curated structural models for each variant from crystal structures and homology models. VIPUR can be applied to mutations in any organism''s proteome with improved generalized accuracy (AUROC .83) and interpretability (AUPR .87) compared to other methods. We demonstrate that VIPUR''s predictions of deleteriousness match the biological phenotypes in ClinVar and provide a clear ranking of prediction confidence. We use VIPUR to interpret known mutations associated with inflammation and diabetes, demonstrating the structural diversity of disrupted functional sites and improved interpretation of mutations associated with human diseases. Lastly, we demonstrate VIPUR''s ability to highlight candidate variants associated with human diseases by applying VIPUR to de novo variants associated with autism spectrum disorders.     

Limitations of Ab Initio Predictions of Peptide Binding to MHC Class II Molecules

Successful predictions of peptide MHC binding typically require a large set of binding data for the specific MHC molecule that is examined. Structure based prediction methods promise to circumvent this requirement by evaluating the physical contacts a peptide can make with an MHC molecule based on the highly conserved 3D structure of peptide:MHC complexes. While several such methods have been described before, most are not publicly available and have not been independently tested for their performance. We here implemented and evaluated three prediction methods for MHC class II molecules: statistical potentials derived from the analysis of known protein structures; energetic evaluation of different peptide snapshots in a molecular dynamics simulation; and direct analysis of contacts made in known 3D structures of peptide:MHC complexes. These methods are ab initio in that they require structural data of the MHC molecule examined, but no specific peptide:MHC binding data. Moreover, these methods retain the ability to make predictions in a sufficiently short time scale to be useful in a real world application, such as screening a whole proteome for candidate binding peptides. A rigorous evaluation of each methods prediction performance showed that these are significantly better than random, but still substantially lower than the best performing sequence based class II prediction methods available. While the approaches presented here were developed independently, we have chosen to present our results together in order to support the notion that generating structure based predictions of peptide:MHC binding without using binding data is unlikely to give satisfactory results.     

Dynameomics: Data‐driven methods and models for utilizing large‐scale protein structure repositories for improving fragment‐based loop prediction

Protein function is intimately linked to protein structure and dynamics yet experimentally determined structures frequently omit regions within a protein due to indeterminate data, which is often due protein dynamics. We propose that atomistic molecular dynamics simulations provide a diverse sampling of biologically relevant structures for these missing segments (and beyond) to improve structural modeling and structure prediction. Here we make use of the Dynameomics data warehouse, which contains simulations of representatives of essentially all known protein folds. We developed novel computational methods to efficiently identify, rank and retrieve small peptide structures, or fragments, from this database. We also created a novel data model to analyze and compare large repositories of structural data, such as contained within the Protein Data Bank and the Dynameomics data warehouse. Our evaluation compares these structural repositories for improving loop predictions and analyzes the utility of our methods and models. Using a standard set of loop structures, containing 510 loops, 30 for each loop length from 4 to 20 residues, we find that the inclusion of Dynameomics structures in fragment‐based methods improves the quality of the loop predictions without being dependent on sequence homology. Depending on loop length, ～25–75% of the best predictions came from the Dynameomics set, resulting in lower main chain root‐mean‐square deviations for all fragment lengths using the combined fragment library. We also provide specific cases where Dynameomics fragments provide better predictions for NMR loop structures than fragments from crystal structures. Online access to these fragment libraries is available at http://www.dynameomics.org/fragments .     

Can contact potentials reliably predict stability of proteins?

The simplest approximation of interaction potential between amino acid residues in proteins is the contact potential, which defines the effective free energy of a protein conformation by a set of amino acid contacts formed in this conformation. Finding a contact potential capable of predicting free energies of protein states across a variety of protein families will aid protein folding and engineering in silico on a computationally tractable time-scale. We test the ability of contact potentials to accurately and transferably (across various protein families) predict stability changes of proteins upon mutations. We develop a new methodology to determine the contact potentials in proteins from experimental measurements of changes in protein's thermodynamic stabilities (DeltaDeltaG) upon mutations. We apply our methodology to derive sets of contact interaction parameters for a hierarchy of interaction models including solvation and multi-body contact parameters. We test how well our models reproduce experimental measurements by statistical tests. We evaluate the maximum accuracy of predictions obtained by using contact potentials and the correlation between parameters derived from different data-sets of experimental (DeltaDeltaG) values. We argue that it is impossible to reach experimental accuracy and derive fully transferable contact parameters using the contact models of potentials. However, contact parameters may yield reliable predictions of DeltaDeltaG for datasets of mutations confined to the same amino acid positions in the sequence of a single protein.     

FREAD revisited: Accurate loop structure prediction using a database search algorithm

Loops are the most variable regions of protein structure and are, in general, the least accurately predicted. Their prediction has been approached in two ways, ab initio and database search. In recent years, it has been thought that ab initio methods are more powerful. In light of the continued rapid expansion in the number of known protein structures, we have re‐evaluated FREAD, a database search method and demonstrate that the power of database search methods may have been underestimated. We found that sequence similarity as quantified by environment specific substitution scores can be used to significantly improve prediction. In fact, FREAD performs appreciably better for an identifiable subset of loops (two thirds of shorter loops and half of the longer loops tested) than the ab initio methods of MODELLER, PLOP, and RAPPER. Within this subset, FREAD's predictive ability is length independent, in general, producing results within 2Å RMSD, compared to an average of over 10Å for loop length 20 for any of the other tested methods. We also benchmarked the prediction protocols on a set of 212 loops from the model structures in CASP 7 and 8. An extended version of FREAD is able to make predictions for 127 of these, it gives the best prediction of the methods tested in 61 of these cases. In examining FREAD's ability to predict in the model environment, we found that whole structure quality did not affect the quality of loop predictions. Proteins 2010. © 2009 Wiley‐Liss, Inc.     

Prediction of protein stability changes for single-site mutations using support vector machines

Accurate prediction of protein stability changes resulting from single amino acid mutations is important for understanding protein structures and designing new proteins. We use support vector machines to predict protein stability changes for single amino acid mutations leveraging both sequence and structural information. We evaluate our approach using cross-validation methods on a large dataset of single amino acid mutations. When only the sign of the stability changes is considered, the predictive method achieves 84% accuracy-a significant improvement over previously published results. Moreover, the experimental results show that the prediction accuracy obtained using sequence alone is close to the accuracy obtained using tertiary structure information. Because our method can accurately predict protein stability changes using primary sequence information only, it is applicable to many situations where the tertiary structure is unknown, overcoming a major limitation of previous methods which require tertiary information. The web server for predictions of protein stability changes upon mutations (MUpro), software, and datasets are available at http://www.igb.uci.edu/servers/servers.html.     

Conditional prediction of consecutive tumor evolution using cancer progression models: What genotype comes next?

Accurate prediction of tumor progression is key for adaptive therapy and precision medicine. Cancer progression models (CPMs) can be used to infer dependencies in mutation accumulation from cross-sectional data and provide predictions of tumor progression paths. However, their performance when predicting complete evolutionary trajectories is limited by violations of assumptions and the size of available data sets. Instead of predicting full tumor progression paths, here we focus on short-term predictions, more relevant for diagnostic and therapeutic purposes. We examine whether five distinct CPMs can be used to answer the question “Given that a genotype with n mutations has been observed, what genotype with n + 1 mutations is next in the path of tumor progression?” or, shortly, “What genotype comes next?”. Using simulated data we find that under specific combinations of genotype and fitness landscape characteristics CPMs can provide predictions of short-term evolution that closely match the true probabilities, and that some genotype characteristics can be much more relevant than global features. Application of these methods to 25 cancer data sets shows that their use is hampered by a lack of information needed to make principled decisions about method choice. Fruitful use of these methods for short-term predictions requires adapting method’s use to local genotype characteristics and obtaining reliable indicators of performance; it will also be necessary to clarify the interpretation of the method’s results when key assumptions do not hold.     

Analysis of ftsZ mutations that confer resistance to the cell division inhibitor SulA (SfiA).

In Escherichia coli, the ftsZ gene is thought to be an essential cell division gene. Several dominant mutations that make lon mutant cells refractory to the cell division inhibitor SulA, sulB9, sulB25, and sfiB114, have been mapped to the ftsZ gene. DNA sequence analysis of these mutations and the sfiB103 mutation confirmed that all of these mutations mapped within the ftsZ gene and revealed that the two sulB mutations were identical and by selection for resistance to higher levels of SulA, contained a second mutation within the ftsZ gene. We therefore propose that these mutations be redesignated ftsZ(Rsa) for resistance to SulA. A procedure involving mutagenesis of ftsZ cloned on low-copy-number vectors was used to isolate three additional ftsZ(Rsa) mutations. DNA sequence analysis of these mutations revealed that they were distinct from the previously isolated mutations. One of these mutations, ftsZ3(Rsa), led to an altered FtsZ protein that could no longer support cell growth but still conferred the Rsa phenotype in the presence of ftsZ+. In addition to being resistant to SulA, all ftsZ(Rsa) mutations also conferred resistance to a LacZ-FtsZ hybrid protein (ZZ). One possibility is that FtsZ functions as a multimer and that FtsZ(Rsa) mutant proteins have an increased ability for multimerization, making them resistant to SulA and ZZ.     

A community resource benchmarking predictions of peptide binding to MHC-I molecules

Recognition of peptides bound to major histocompatibility complex (MHC) class I molecules by T lymphocytes is an essential part of immune surveillance. Each MHC allele has a characteristic peptide binding preference, which can be captured in prediction algorithms, allowing for the rapid scan of entire pathogen proteomes for peptide likely to bind MHC. Here we make public a large set of 48,828 quantitative peptide-binding affinity measurements relating to 48 different mouse, human, macaque, and chimpanzee MHC class I alleles. We use this data to establish a set of benchmark predictions with one neural network method and two matrix-based prediction methods extensively utilized in our groups. In general, the neural network outperforms the matrix-based predictions mainly due to its ability to generalize even on a small amount of data. We also retrieved predictions from tools publicly available on the internet. While differences in the data used to generate these predictions hamper direct comparisons, we do conclude that tools based on combinatorial peptide libraries perform remarkably well. The transparent prediction evaluation on this dataset provides tool developers with a benchmark for comparison of newly developed prediction methods. In addition, to generate and evaluate our own prediction methods, we have established an easily extensible web-based prediction framework that allows automated side-by-side comparisons of prediction methods implemented by experts. This is an advance over the current practice of tool developers having to generate reference predictions themselves, which can lead to underestimating the performance of prediction methods they are not as familiar with as their own. The overall goal of this effort is to provide a transparent prediction evaluation allowing bioinformaticians to identify promising features of prediction methods and providing guidance to immunologists regarding the reliability of prediction tools.     

An efficient strategy for extensive integration of diverse biological data for protein function prediction

MOTIVATION: With the increasing availability of diverse biological information, protein function prediction approaches have converged towards integration of heterogeneous data. Many adapted existing techniques, such as machine-learning and probabilistic methods, which have proven successful on specific data types. However, the impact of these approaches is hindered by a couple of factors. First, there is little comparison between existing approaches. This is in part due to a divergence in the focus adopted by different works, which makes comparison difficult or even fuzzy. Second, there seems to be over-emphasis on the use of computationally demanding machine-learning methods, which runs counter to the surge in biological data. Analogous to the success of BLAST for sequence homology search, we believe that the ability to tap escalating quantity, quality and diversity of biological data is crucial to the success of automated function prediction as a useful instrument for the advancement of proteomic research. We address these problems by: (1) providing useful comparison between some prominent methods; (2) proposing Integrated Weighted Averaging (IWA)--a scalable, efficient and flexible function prediction framework that integrates diverse information using simple weighting strategies and a local prediction method. The simplicity of the approach makes it possible to make predictions based on on-the-fly information fusion. RESULTS: In addition to its greater efficiency, IWA performs exceptionally well against existing approaches. In the presence of cross-genome information, which is overwhelming for existing approaches, IWA makes even better predictions. We also demonstrate the significance of appropriate weighting strategies in data integration.     

MetAmyl: A METa-Predictor for AMYLoid Proteins

The aggregation of proteins or peptides in amyloid fibrils is associated with a number of clinical disorders, including Alzheimer''s, Huntington''s and prion diseases, medullary thyroid cancer, renal and cardiac amyloidosis. Despite extensive studies, the molecular mechanisms underlying the initiation of fibril formation remain largely unknown. Several lines of evidence revealed that short amino-acid segments (hot spots), located in amyloid precursor proteins act as seeds for fibril elongation. Therefore, hot spots are potential targets for diagnostic/therapeutic applications, and a current challenge in bioinformatics is the development of methods to accurately predict hot spots from protein sequences. In this paper, we combined existing methods into a meta-predictor for hot spots prediction, called MetAmyl for METapredictor for AMYLoid proteins. MetAmyl is based on a logistic regression model that aims at weighting predictions from a set of popular algorithms, statistically selected as being the most informative and complementary predictors. We evaluated the performances of MetAmyl through a large scale comparative study based on three independent datasets and thus demonstrated its ability to differentiate between amyloidogenic and non-amyloidogenic polypeptides. Compared to 9 other methods, MetAmyl provides significant improvement in prediction on studied datasets. We further show that MetAmyl is efficient to highlight the effect of point mutations involved in human amyloidosis, so we suggest this program should be a useful complementary tool for the diagnosis of these diseases.     

A Gene-Specific Method for Predicting Hemophilia-Causing Point Mutations

A fundamental goal of medical genetics is the accurate prediction of genotype–phenotype correlations. As an approach to develop more accurate in silico tools for prediction of disease-causing mutations of structural proteins, we present a gene- and disease-specific prediction tool based on a large systematic analysis of missense mutations from hemophilia A (HA) patients. Our HA-specific prediction tool, HApredictor, showed disease prediction accuracy comparable to other publicly available prediction software. In contrast to those methods, its performance is not limited to non-synonymous mutations. Given the role of synonymous mutations in disease and drug codon optimization, we propose that utilizing a gene- and disease-specific method can be highly useful to make functional predictions possible even for synonymous mutations. Incorporating computational metrics at both nucleotide and amino acid levels along with multiple protein sequence/structure alignment significantly improved the predictive performance of our tool. HApredictor is freely available for download at http://www.ncbi.nlm.nih.gov/CBBresearch/Przytycka/HA_Predict/index.htm.     

Using Knowledge of Protein Structural Constraints to Predict the Evolution of HIV-1

The high levels of sequence diversity and rapid rates of evolution of HIV-1 represent the main challenges for developing effective therapies. However, there are constraints imposed by the three-dimensional protein structure that affect the sequence space accessible to the evolution of HIV-1. Here, we present a strategy for predicting the set of possible amino acid replacements in HIV. Our approach is based on the identification of likely amino acid changes in the context of these structural constraints using environment-specific substitution matrices as well as considering the physical constraints imposed by local structure. Assessment of the power of various published algorithms in predicting the evolution of HIV-1 Gag P17 shows that it is possible to use these methods to make accurate predictions of the sequence diversity. Our own method, SubFit, uses knowledge of local structural constraints; it achieves similar prediction success with the best-performing methods. We also show that erroneous predictions are largely due to infrequently occurring amino acids that will probably have severe fitness costs for the protein. Future improvements; for example, incorporating covariation and immunological constraints will permit more reliable prediction of viral evolution.     

FISHER'S GEOMETRIC MODEL OF ADAPTATION MEETS THE FUNCTIONAL SYNTHESIS: DATA ON PAIRWISE EPISTASIS FOR FITNESS YIELDS INSIGHTS INTO THE SHAPE AND SIZE OF PHENOTYPE SPACE

The functional synthesis uses experimental methods from molecular biology, biochemistry and structural biology to decompose evolutionarily important mutations into their more proximal mechanistic determinants. However these methods are technically challenging and expensive. Noting strong formal parallels between R.A. Fisher's geometric model of adaptation and a recent model for the phenotypic basis of protein evolution, we sought to use the former to make inferences into the latter using data on pairwise fitness epistasis between mutations. We present an analytic framework for classifying pairs of mutations with respect to similarity of underlying mechanism on this basis, and also show that these data can yield an estimate of the number of mutationally labile phenotypes underlying fitness effects. We use computer simulations to explore the robustness of our approach to violations of analytic assumptions and analyze several recently published datasets. This work provides a theoretical complement to the functional synthesis as well as a novel test of Fisher's geometric model.     

Prediction of protein-protein interactions: the CAPRI experiment, its evaluation and implications

Given the increasing interest in protein-protein interactions, the prediction of these interactions from sequence and structural information has become a booming activity. CAPRI, the community-wide experiment for assessing blind predictions of protein-protein interactions, is playing an important role in fostering progress in docking procedures. At the same time, novel methods are being derived for predicting regions of a protein that are likely to interact and for characterizing putative intermolecular contacts from sequence and structural data. Together with docking procedures, these methods provide an integrated computational approach that should be a valuable complement to genome-scale experimental studies of protein-protein interactions.